{"id":"https://openalex.org/W4416657212","doi":"https://doi.org/10.1007/s10994-025-06928-5","title":"Interpretable Deep Reinforcement Learning Via Concept-Based Policy Distillation","display_name":"Interpretable Deep Reinforcement Learning Via Concept-Based Policy Distillation","publication_year":2025,"publication_date":"2025-11-25","ids":{"openalex":"https://openalex.org/W4416657212","doi":"https://doi.org/10.1007/s10994-025-06928-5"},"language":"en","primary_location":{"id":"doi:10.1007/s10994-025-06928-5","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10994-025-06928-5","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10994-025-06928-5.pdf","source":{"id":"https://openalex.org/S62148650","display_name":"Machine Learning","issn_l":"0885-6125","issn":["0885-6125","1573-0565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Machine Learning","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s10994-025-06928-5.pdf","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5001301343","display_name":"Yanzhe Bekkemoen","orcid":"https://orcid.org/0000-0001-6487-6149"},"institutions":[{"id":"https://openalex.org/I204778367","display_name":"Norwegian University of Science and Technology","ror":"https://ror.org/05xg72x27","country_code":"NO","type":"education","lineage":["https://openalex.org/I204778367"]}],"countries":["NO"],"is_corresponding":true,"raw_author_name":"Yanzhe Bekkemoen","raw_affiliation_strings":["Department of Computer Science, Norwegian University of Science and Technology, 7491, Trondheim, Norway"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Norwegian University of Science and Technology, 7491, Trondheim, Norway","institution_ids":["https://openalex.org/I204778367"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5040301804","display_name":"Helge Langseth","orcid":"https://orcid.org/0000-0001-6324-6284"},"institutions":[{"id":"https://openalex.org/I204778367","display_name":"Norwegian University of Science and Technology","ror":"https://ror.org/05xg72x27","country_code":"NO","type":"education","lineage":["https://openalex.org/I204778367"]}],"countries":["NO"],"is_corresponding":false,"raw_author_name":"Helge Langseth","raw_affiliation_strings":["Department of Computer Science, Norwegian University of Science and Technology, 7491, Trondheim, Norway"],"affiliations":[{"raw_affiliation_string":"Department of Computer Science, Norwegian University of Science and Technology, 7491, Trondheim, Norway","institution_ids":["https://openalex.org/I204778367"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5001301343"],"corresponding_institution_ids":["https://openalex.org/I204778367"],"apc_list":{"value":2390,"currency":"EUR","value_usd":2990},"apc_paid":{"value":2390,"currency":"EUR","value_usd":2990},"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.19761856,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"114","issue":"12","first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.6672999858856201,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.6672999858856201,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.08429999649524689,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.027799999341368675,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6392999887466431},{"id":"https://openalex.org/keywords/convolutional-neural-network","display_name":"Convolutional neural network","score":0.6144000291824341},{"id":"https://openalex.org/keywords/hyperparameter","display_name":"Hyperparameter","score":0.5981000065803528},{"id":"https://openalex.org/keywords/black-box","display_name":"Black box","score":0.5408999919891357},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5062000155448914},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.4674000144004822},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.41510000824928284},{"id":"https://openalex.org/keywords/matrix-decomposition","display_name":"Matrix decomposition","score":0.37369999289512634}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6969000101089478},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6392999887466431},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6259999871253967},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.6144000291824341},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.5981000065803528},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5494999885559082},{"id":"https://openalex.org/C94966114","wikidata":"https://www.wikidata.org/wiki/Q29256","display_name":"Black box","level":2,"score":0.5408999919891357},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5062000155448914},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4674000144004822},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.41510000824928284},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.37369999289512634},{"id":"https://openalex.org/C163175372","wikidata":"https://www.wikidata.org/wiki/Q3339222","display_name":"Linear model","level":2,"score":0.3659000098705292},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.35569998621940613},{"id":"https://openalex.org/C75294576","wikidata":"https://www.wikidata.org/wiki/Q5165192","display_name":"Contextual image classification","level":3,"score":0.3440000116825104},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.3384999930858612},{"id":"https://openalex.org/C123587114","wikidata":"https://www.wikidata.org/wiki/Q2101508","display_name":"Policy analysis","level":2,"score":0.32659998536109924},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2937000095844269},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.27559998631477356},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.274399995803833},{"id":"https://openalex.org/C154030694","wikidata":"https://www.wikidata.org/wiki/Q1436074","display_name":"Fractionating column","level":3,"score":0.25859999656677246},{"id":"https://openalex.org/C56372850","wikidata":"https://www.wikidata.org/wiki/Q1050404","display_name":"Sparse matrix","level":3,"score":0.25540000200271606}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1007/s10994-025-06928-5","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10994-025-06928-5","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10994-025-06928-5.pdf","source":{"id":"https://openalex.org/S62148650","display_name":"Machine Learning","issn_l":"0885-6125","issn":["0885-6125","1573-0565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Machine Learning","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1007/s10994-025-06928-5","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s10994-025-06928-5","pdf_url":"https://link.springer.com/content/pdf/10.1007/s10994-025-06928-5.pdf","source":{"id":"https://openalex.org/S62148650","display_name":"Machine Learning","issn_l":"0885-6125","issn":["0885-6125","1573-0565"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Machine Learning","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320311040","display_name":"Norges Teknisk-Naturvitenskapelige Universitet","ror":"https://ror.org/05xg72x27"},{"id":"https://openalex.org/F4320327267","display_name":"St. Olavs Hospital Universitetssykehuset i Trondheim","ror":"https://ror.org/01a4hbq44"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416657212.pdf","grobid_xml":"https://content.openalex.org/works/W4416657212.grobid-xml"},"referenced_works_count":48,"referenced_works":["https://openalex.org/W1849277567","https://openalex.org/W1902027874","https://openalex.org/W2011301426","https://openalex.org/W2150468603","https://openalex.org/W2242818861","https://openalex.org/W2257979135","https://openalex.org/W2271840356","https://openalex.org/W2594227402","https://openalex.org/W2606462007","https://openalex.org/W2736601468","https://openalex.org/W2741375793","https://openalex.org/W2787894218","https://openalex.org/W2897798332","https://openalex.org/W2962790223","https://openalex.org/W2963598268","https://openalex.org/W2964121744","https://openalex.org/W2966831185","https://openalex.org/W2971538581","https://openalex.org/W2989847975","https://openalex.org/W3003257820","https://openalex.org/W3035965352","https://openalex.org/W3082925502","https://openalex.org/W3098338892","https://openalex.org/W3116073702","https://openalex.org/W3124922852","https://openalex.org/W3127493833","https://openalex.org/W3127561923","https://openalex.org/W3137125108","https://openalex.org/W3157950068","https://openalex.org/W3172917901","https://openalex.org/W3204942516","https://openalex.org/W4220696590","https://openalex.org/W4298857966","https://openalex.org/W4307359761","https://openalex.org/W4386076059","https://openalex.org/W4386193536","https://openalex.org/W4386741020","https://openalex.org/W4388230841","https://openalex.org/W4389131286","https://openalex.org/W4389607308","https://openalex.org/W4390874575","https://openalex.org/W4391897126","https://openalex.org/W4391924614","https://openalex.org/W4392964510","https://openalex.org/W4394963598","https://openalex.org/W4404738510","https://openalex.org/W6912515347","https://openalex.org/W6968906521"],"related_works":[],"abstract_inverted_index":{"Abstract":[0],"Deep":[1],"reinforcement":[2],"learning":[3],"policies":[4,39],"perform":[5,197],"exceptionally":[6],"well":[7,198],"in":[8,40,90,161,217],"applications":[9,42],"like":[10,43],"Atari":[11],"games,":[12],"chess,":[13],"Go,":[14],"and":[15,28,46,94,168,180,210,225],"poker.":[16],"However,":[17],"they":[18],"are":[19,231],"incomprehensible,":[20],"making":[21,143],"the":[22,34,53,83,97,100,105,141,155,185,202,206,212],"process":[23],"of":[24,127,157],"extracting":[25],"new":[26,59],"knowledge":[27],"understanding":[29],"policy":[30,61,98,111,119,150,160,188,209],"behavior":[31,216],"difficult.":[32],"For":[33],"same":[35,203],"reason,":[36],"deploying":[37],"these":[38,175,194],"high-stakes":[41],"healthcare,":[44],"finance,":[45],"criminal":[47],"justice":[48],"is":[49],"infeasible.":[50],"To":[51],"rectify":[52],"incomprehensibility":[54],"issue,":[55],"we":[56,153,191],"propose":[57],"a":[58,109,125,135],"concept-based":[60],"distillation":[62],"method":[63,70],"for":[64,146],"convolutional":[65],"neural":[66],"network-based":[67],"policies.":[68],"Our":[69],"transforms":[71],"raw":[72],"image":[73],"states":[74],"into":[75,183],"human-interpretable":[76],"concepts":[77,87,106],"using":[78,113,201],"non-negative":[79],"matrix":[80],"factorization":[81],"on":[82],"policy\u2019s":[84,215],"activations.":[85],"The":[86,117,220],"express":[88],"features":[89,204],"an":[91],"interpretable":[92],"way":[93],"detail":[95],"how":[96,184],"represents":[99],"world":[101],"internally.":[102],"We":[103,171],"use":[104],"to":[107,130,148],"train":[108],"distilled":[110,118,159],"represented":[112],"sparse":[114,137],"linear":[115,122,128,138,176,195],"models.":[116],"chooses":[120],"one":[121],"model":[123,139],"from":[124],"set":[126],"models":[129,177,196],"make":[131],"action":[132],"predictions.":[133],"Employing":[134],"single":[136],"reduces":[140],"complexity,":[142],"it":[144],"easier":[145],"humans":[147],"understand":[149],"behavior.":[151],"Experimentally,":[152],"show":[154],"effectiveness":[156],"our":[158],"four":[162],"environments:":[163],"Car":[164],"Racing,":[165],"Pong,":[166],"Breakout,":[167],"Ms":[169],"Pacman.":[170],"illustrate":[172],"that":[173,193],"inspecting":[174],"gives":[178],"local":[179],"global":[181],"insight":[182],"black":[186,207,213],"box":[187,208,214],"works.":[189],"Furthermore,":[190],"demonstrate":[192],"by":[199],"faithfully":[200],"as":[205],"capturing":[211],"critical":[218],"states.":[219],"code,":[221],"data,":[222],"trained":[223],"models,":[224],"TensorBoard":[226],"logs":[227],"with":[228],"hyperparameters":[229],"used":[230],"provided":[232],"(":[233],"https://github.com/observer4599/interpretable-concept-based-policy-distillation":[234],").":[235]},"counts_by_year":[],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-11-27T00:00:00"}